Radeon Device aa68 Graphics card (Cedar). Used in hybrid graphics with Intel i915 vgaswitcheroo, hence both are compiled in (Bug #36783 ) and KMS enabled (without KMS the problem usually not occurs). On boot, the screen tunrs of the backlight by itsself and has to be turned on manually. A message in dmesg that is likely to be related says
Firmware Bug: ACPI: No _BCQ method
THe log has several errors related to the radeon. Kernel is 39-rc5.
Created attachment 46307 [details]
Created attachment 46384 [details] [review]
Does this kernel patch help?
(In reply to comment #2)
> Created an attachment (id=46384) [details]
> possible fix
> Does this kernel patch help?
No (tried on next-20110506).
This looks like a duplicate of bug 41306.
(In reply to comment #4)
> This looks like a duplicate of bug 41306.
Negative. It is related in thus far as that this problem (as I've mentioned in Bug #41306) constitutes the part where the backlight initially goes off. However, it's is not assumed, nor it is likely, that it is directly related to the problem that the backlight cannot be brought back on (which is, what the other bug concerns).
Prior to the 3 series there have been times when the backlight would go off (with the error mentioned above) but could be brought back on nonetheless.
As also mentioned in the other bug, these scenarios could, however, not be reproduced for unknown reasons.
After all this bug should stay open and a priori assumed to be uncorrelated.